An Unsupervised Bayesian Modelling Approach for Storyline Detection on News Articles

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1943–1948, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. An Unsupervised Bayesian Modelling Approach to Storyline Detection from News Articles Deyu Zhou†‡ Haiyang Xu† Yulan He§ † School of Computer Science and Engineering, Southeast University, China ‡ State Key Laboratory for Novel Software Technology, Nanjing University, China § School of Engineering and Applied Science, Aston University, UK d.zhou@seu.edu.cn, h.xu@seu.edu.cn, y.he@cantab.net Abstract Storyline detection from news articles aims at summarizing events described un- der a certain news topic and revealing how those events evolve over time. It is a dif- ficult task because it requires first the de- tection of events from news articles pub- lished in different time periods and then the construction of storylines by linking events into coherent news stories. More- over, each storyline has different hierarchi- cal structures which are dependent across epochs. Existing approaches often ignore the dependency of hierarchical structures in storyline generation. In this paper, we propose an unsupervised Bayesian model, called dynamic storyline detection model, to extract structured representations and evolution patterns of storylines. The pro- posed model is evaluated on a large scale news corpus. Experimental results show that our proposed model outperforms sev- eral baseline approaches. 1 Introduction The rapid development of online news media sites is accompanied by the generation of tremendous news reports. Facing such massive amount of news articles, it is crucial to develop an automat- ed tool which can provide a temporal summary of events and their evolutions related to a topic from news reports. Therefore, storyline detection, aim- ing at summarising the development of certain re- lated events, has been studied in order to help read- ers quickly understand the major events reported in news articles. It has attracted great attention re- cently. Kawamae (2011) proposed a trend analy- sis model which used the difference between tem- poral words and other words in each documen- t to detect topic evolution over time. Ahmed et al. (2011) proposed a unified framework to group temporally and topically related news articles in- to same storylines in order to reveal the tempo- ral evolution of events. Tang and Yang (2012) de- veloped a topic-user-trend model, which incorpo- rates user interests into the generative process of web contents. Radinsky and Horvitz (2013) built storylines based on text clustering and entity en- tropy to predict future events. Huang and Huang (2013) developed a mixture-event-aspect model to model sub-events into local and global aspects and utilize an optimization method to generate story- lines. Wang et al. (2013) proposed an evolutionary multi-branch tree clustering method for streaming text data in which the tree construction is casted as an online posterior estimation problem by con- sidering both the current tree and the previous tree simultaneously. With the fast development of social media plat- forms, newsworthy events are widely scattered not only on traditional news media but also on social media (Zhou et al., 2015). For example, Twit- ter, one of …

Content truncated to 500 words